Biophysical Chemistry 111 (2004) 43 – 52

www.elsevier.com/locate/bpc

ESR studies on the effect of cholesterol on chlorpromazine interaction

with saturated and unsaturated liposome membranes
Anna Wisniewska *, Agnieszka Wolnicka-Glubisz 1
Department of Biophysics, Faculty of Biotechnology, Jagiellonian University, Gronostajowa 7, 30-387 Cracow, Poland

Received in revised form 12 April 2004; accepted 14 April 2004

Available online 6 May 2004

Abstract

In this study, the effects of chlorpromazine (CPZ) on lipid order and motion in saturated (DMPC, DMPG) and unsaturated (SOPC)
liposome membranes were investigated by electron spin resonance (ESR) spin labeling technique. We have shown that above the main phase
transition temperature of membrane lipids (TM), CPZ slightly increases lipid order in membranes without cholesterol, whereas below TM it
has a strong opposite effect. Addition of 30 mol% of cholesterol into DMPC and SOPC membranes changes significantly the CPZ effects
both above and below TM. Additionally, above TM, the ordering effect of CPZ on pure SOPC membrane is stronger at pH 7.4 than at pH 9.0,
whereas below TM, as well as in the presence of cholesterol, pH does not seem to play a role in CPZ effect on both membranes. Because of
the strong influence of membrane composition on CPZ effect on membranes, the use of cholesterol as a marker of CPZ photosensitized
reactions has been discussed.
D 2004 Elsevier B.V. All rights reserved.

Keywords: Chlorpromazine; Phospholipid membrane; Cholesterol; Spin label; ESR

1. Introduction 6], and perturbs the membrane structure. CPZ, as a basic

Chlorpromazine (CPZ) is a neuroleptic drug of the
phenothiazine group, widely used in the treatment of certain
psychiatric disorders. Its amphipathic character enables it to
partition into the lipid bilayer of cell membranes, and to
reach in this way the central nervous system. Interactions of
CPZ with biological membranes are linked to the numerous
effects in cellular systems, such as a well-known protective
effect on red cells osmolysis [1] and redirecting de novo
synthesis from neutral to acidic glycerophospholipids by
inhibition of PA phosphatase in hepatocytes [2]. Further-
more, it has been shown that neuroleptic drugs affect
the function of various receptors including adrenergic,
muscarinic, histamine, 5-hydroxytryptamine and dopamine
receptor families [3]. Numerous experiments on model
membranes have shown that CPZ interacts with membrane
phospholipids, especially with negatively charged ones [4 –
* Corresponding author. Tel.: +48-12-664-63-55; fax: +48-12-664-
69-02.
E-mail address: wisnia@mol.uj.edu.pl (A. Wisniewska).
1
Present address: Department of Environmental and Occupational
Health, School of Public Health and Health Services, The George
Washington University, Medical Center, Washington, DC, USA.
molecule with pK = 8.6 [7], remains mostly protonated and
positively charged at physiological pH, which makes it
locate preferably in the membrane – water interface [8]. Its
binding sites (such as negatively charged carboxyl groups)
are localized near the surface of the membrane, as shown by
the ESR study of Louro et al. [9]. Some data, however,
suggest that CPZ may also penetrate into the acyl chain
region of phospholipid membranes affecting the acyl chain
order [10] and lipid phase transition [4,11]. It has been
shown that the partition coefficient (kp) of CPZ into the
lipid membrane decreases with the increasing phospholipid
alkyl chain length [12], and can be additionally reduced by
the presence of cholesterol [12,13].
The understanding of CPZ localization and behavior in
membranes is interesting also from the photochemical point
of view, since its application combined with sunlight expo-
sure can cause side effects including numerous ocular
complications [14] and photoalergy [15,16]. Photochemistry
of CPZ is rather complicated and strongly depends on
wavelength [17,18]. Although studies by Ljunggren and
Moller [19] and Schoonderwoerd et al. [20] indicated that
dechloration is responsible for phototoxicity of CPZ, free
radicals and singlet oxygen (1O2) also appear to play an

0301-4622/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.bpc.2004.04.001
44 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52
important role in CPZ photosensitization. Interestingly, CPZ
photogenerates 1O2 in organic solvents, whereas in aqueous
solutions does not [21]. The interaction of CPZ with mem-
branes and the extent of its partition into the membrane is
then important, since CPZ may play a role of a photosensi-
tizer in the membrane interior by being able to generate 1O2
in its hydrophobic surroundings. Additionally, unsaturated
phospholipids and cholesterol in cell membranes are suscep-
tible to reactive oxygen species being generated during
photosensitized reactions. Like phospholipids, cholesterol
is an important constituent of animal membranes, where it
comprises 40 – 45 mol% of the total lipid, but because it
exists as a single molecular species in natural membranes, it
presents much fewer oxidation products. In cells under
oxidative stress, membranous cholesterol, through the for-
mation of its signature hydroperoxide and diol products
serves as a unique marker in situ, allowing discrimination
between 1O2 and free radical intermediacy [22 – 24].
On the other hand, it is well known that cholesterol
strongly affects various membrane properties, among which
its effects on membrane fluidity, phase transition and pene-
tration of polar and non-polar molecules have been most
extensively studied [25 –28]. Also, it has been shown that the
presence of unsaturated double bonds in the phospholipid
alkyl chain strongly affects membrane properties [27,29] and
the penetration of different drugs, such as tetracycline [30].
We have been therefore interested in two main problems: (1)
how does the membrane composition (especially lipid alkyl
chain unsaturation and the presence of cholesterol) influence
the CPZ effect on membranes, and (2) how relevant is the use
of cholesterol as a marker of CPZ photosensitized reactions
in case of altered membrane properties.
To address these issues we have undertaken an investi-
gation of the CPZ effect on lipid order and motion in
different liposome membranes. The factors analyzed in the
present work were (1) the presence of unsaturated double
bonds in lipid alkyl chains; (2) the presence of cholesterol;
(3) temperature (as a factor causing the membrane phase
transition); (4) pH (affecting CPZ charge); and (5) the
surface charge of the membrane.
2. Materials and methods
2.1. Preparation of liposomes
L-a phosphatidylcholine, dimyristoyl (DMPC), L-a phos-
phatidyl-DL-glycerol, dimyristoyl (DMPG), L-a phosphati-
dylcholine-h-oleoyl-g-stearyoyl (SOPC) and cholesterol
were purchased from Sigma (Germany), L-a phosphatidyl-
choline-h-oleoyl-g-palmitoyl (POPC) from Avanti Polar
Lipids (USA), CPZ from Sigma Aldrich (Belgium), and
spin labels from Molecular Probes (USA). The membranes
used in this work were multilamellar liposomes containing 0
or 30 mol% of cholesterol; 0, 10 or 30 mol% of CPZ and 1
mol% of spin label (in case of ESR measurements), pre-
pared according to Ref. [25]. All compounds were dissolved
in chloroform, which was subsequently evaporated under
stream of nitrogen to form a lipid film, which was then put
under vacuum for at least 14 h. The dried lipids were
suspended either in 0.1 M PBS (pH 7.4) or in 0.1 M borate
buffer (pH 9.0) and vortexed vigorously. The multilamellar
liposome suspension was centrifuged at 7000 rpm, for 15
min at 4 jC and the pellet (about 30 Al) was used as the
sample for ESR measurements.
2.2. ESR measurements of lipid bilayer fluidity
The ESR measurements were performed on a Varian E-3
spectrometer at 10 mW of microwave power. Liposome
samples were deoxygenated under argon flow, placed in the
Pasteur pipettes (Medlab Products, Raszyn, Poland; length
230 mm, / = 1 mm) and centered in the resonant cavity. The
measurements were carried out above and below TM of the
lipid membranes (at 37 and 15 jC in case of the saturated
DMPC and DMPG membranes, and at 15 and 2 jC in case of
the unsaturated SOPC membrane). SOPC has been chosen
for ESR measurements because its TM is + 6 jC, whereas
more frequently used unsaturated lipids such as POPC or egg
yolk PC (EYPC) have TM well below 0 jC [31]. Therefore it
was possible to do measurements below TM of SOPC for still
not frozen samples. ESR spectra of n-doxylstearic acid spin
labels (n-SASL, where n = 5 or 16) or cholestane spin label
(CSL) incorporated into DMPC, DMPG or SOPC liposomes
were analyzed in terms of the order parameter (S), correlation
times and maximum splitting parameter (Amax).
Above TM, to monitor the lipid order and motion the order
parameter (S) was used. It was calculated using the equations
[32]:
S ¼ 0:5407ðTII  T? Þ=ao Þ for n  SASL ð1Þ
or
S ¼ 1:131ðTII  T? Þ=ao Þ for CSL ð2Þ
where
ao ¼ ðTII þ 2T? Þ=3 ð3Þ
and TII and T? were measured directly from the ESR spectra.
S parameter as well as TII and T? values are in principle
static parameters. 16-SASL exhibits however so much
motion that a different analysis can also be used [33]. The
effective correlation times s2B and s2C assuming isotropic
rotational diffusion of 16-SASL, were calculated according
to the formulas [34]:
s2B ¼ 6:51  1010 DHo ½ðho =h Þ1=2  ½ðho =hþ Þ1=2 s ð4Þ
s2C ¼ 6:51  10 10 1=2
DHo ½ðho =h Þ þ ½ðho =hþ Þ 1=2
 2 s
ð5Þ
where: DHo is the peak-to-peak width of the central line in
gauss and h+, ho and h are heights of the low, central and
high field peaks, respectively. When s2B and s2C are similar,
the motion is considered to be isotropic.
 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52
Below TM, a maximum splitting parameter (Amax) mea-
sured directly from the ESR spectra as the separation
between the outer hyperfine lines, was used to monitor the
lipid alkyl chain order [32]. Amax value decreases as mo-
tional freedom increases.
2.3. Spectrophotometric measurements of chlorpromazine
partition into liposomes
The studies were performed on DMPC and POPC
membranes in the presence and absence of 5 mM choles-
terol and at the CPZ concentration of 0.5 mM at pH 7.4.
POPC has been chosen instead of SOPC because at 4 jC,
which is the temperature required to separating the liposome
and buffer phases by centrifugation [25], and at which the
equilibrium between these two phases has been achieved,
POPC is in the liquid-crystalline state (its TM is below zero),
whereas both DMPC and SOPC are in the gel state
(additionally, 4 jC is not a convenient temperature for
SOPC, because it is very close to its TM).. Therefore, using
of POPC gave us an insight into the CPZ partition into
liposome membranes also above TM. The supernatants of
centrifuged multilamellar liposome suspensions were care-
fully collected and transferred to new eppendorf tubes and
their absorbance was measured. All absorption spectra were
recorded on an UV – Vis HP 8452 (Hewlett Packard) diode
spectrophotometer in the range 200 –800 nm with a spectral
resolution 2 nm. The supernatants of CPZ containing lip-
osomes were measured versus the supernatants of
corresponding pure liposomes, whereas the absorption of
CPZ in the buffer was measured versus the buffer alone. The
calculation was done by normalization at k = 308 nm, which
is the third CPZ absorption peak. The absorbance of the
CPZ buffer solution (0.5 mM) was set as 100%. The
percentage of CPZ which incorporated into or associated
with lipid membrane was calculated by comparing the
absorbance of the supernatant with the absorbance of the
CPZ buffer solution. The measurements at pH 9.0 were
impossible due to high scattering effects of the CPZ buffer
solution itself.
2.4. Statistical analysis
The bars represent the mean value, and the errors were
calculated as standard deviation (SD) of at least four
independent measurements. Samples were statistically ana-
lyzed by the Student’s t-test ( p V 0.05).
3. Results
3.1. Comparison of chlorpromazine effects on lipid order
and motion in DMPC and DMPG membranes
The effect of CPZ on lipid order and motion in neutral
and negatively charged membranes was investigated by
45
using three spin labels incorporated into DMPC or DMPG
liposomes at pH 9.0. A sterol-type CSL (with the nitroxide
free radical moiety located very close to the polar head
groups) is a probe reflecting the behavior of the whole rigid
lipid molecules such as cholesterol, whereas 5-SASL and
16-SASL (with the nitroxide moiety close to the polar head
groups and at the membrane center, respectively) are good
probes of the segmental motion of phospholipid alkyl chains
[25]. The ESR spectra were collected both above and below
TM of DMPC and DMPG membranes (at 37 and 15 jC,
respectively).
3.1.1. Measurements above TM
Table 1 shows the values of S parameter calculated from
ESR spectra of CSL, 5-SASL and 16-SASL incorporated
into DMPC and DMPG membranes, in the absence and in
the presence of 10 mol% of CPZ. A slight increase in S
parameter can be seen in the presence of 10 mol% of CPZ,
but the differences between pure membranes and those
containing CPZ are in most cases very small. The ordering
effect of CPZ is very low in all regions of DMPC mem-
brane, unlike in DMPG, where it is stronger for CSL (close
to the polar headgroups).
In Table 2 the effect of 10 mol% of CPZ on rotational
motion of 16-SASL in DMPC and DMPG membranes is
presented. Again, the ordering effect of CPZ (expressed as a
decrease in motional freedom of 16-SASL free radical
moiety) in both membranes is very low, being slightly
stronger for DMPG. In both cases, the addition of 10
mol% of CPZ does not affect the isotropy of the motion.
The stronger effect of CPZ on DMPG liposomes com-
pared to DMPC ones shows that the negative charge at the
membrane surface enhances to some extent the interaction of
CPZ with the membrane and its penetration through it. This
result agrees with most literature data [4,6,35], according to
which CPZ interacts stronger with negatively charged mem-
branes. However, the differences are not significant.
3.1.2. Measurements below TM
Fig. 1 shows the effect of 10 mol% of CPZ on motional
freedom of CSL and SASL incorporated into DMPC and
DMPG membranes at 15 jC. DAmax is a difference
between the Amax value in the presence of 10 mol%
CPZ and the control (unperturbed) value of Amax in pure
liposomes. The figure clearly demonstrates that below TM
Table 1
Order parameter (S) obtained from ESR spectra of CSL, 5-SASL and 16-
SASL incorporated into DMPC and DMPG liposomes at 37 jC, in the
presence and in the absence of 10 mol% of CPZ, at pH 9.0
Lipid Order parameter S of spin labels
CSL 5-SAL 16-SASL
DMPC 0.41 0.57 0.08
DMPC + CPZ 0.42 0.58 0.09
DMPG 0.39 0.54 0.08
DMPG + CPZ 0.46 0.55 0.08
46 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52

Table 2
Correlation times of 16-SASL incorporated into DMPC and DMPG
liposomes at 37 jC, in the presence and in the absence of 10 mol% of CPZ,
at pH 9.0
Lipid Correlation time of 16-SASL (ns)
s2B
DMPC 0.61
DMPC + CPZ 0.64
DMPG 0.67
DMPG + CPZ 0.76
the addition of 10 mol% of CPZ strongly increases
motional freedom of lipid molecules. The effects are
similar in both membranes and much stronger for CSL
than for SASL. This can be explained as follows. First, it
may suggest that a rigid molecule of CSL is more sensitive
to the CPZ effect than lipid alkyl chains of SASL. It was
shown before that the effect of some membrane modifiers,
such as cholesterol or carotenoids, on steroid-type spin
labels (like CSL or ASL) was much larger than their effect
on SASL, because CSL is more sensitive to the sum of
changes at various depths in the membrane, while SASL
reflects the local change [33]. Another explanation for the
observed difference between both types of spin labels can
also be taken into consideration, namely the localization of
their nitroxide groups. If CPZ is present mostly at the
membrane – water interface, its strong effect in this region
may be best probed by the nitroxide free radical moiety of
CSL, which is located very close to the membrane head-
group region.
Because of the strongest effect, CSL has been used in
further experiments on the CPZ-membrane interactions.
Moreover, CSL has no surface charge, unlike n-SASLs,
which are negatively charged at pH used. Therefore the
real interaction of CPZ with DMPC or DMPG has been
investigated, not its interaction with the spin label carboxyl
groups, which was proved to occur [9].
3.2. Effect of chlorpromazine on lipid order and motion in
s2C DMPC and SOPC membranes in the presence of cholesterol
0.59
0.65 It is well known that cholesterol strongly affects mem-
0.69 brane fluidity. At the concentration of 30 mol%, cholesterol
0.78 abolishes the main phase transition rigidifying membranes
above TM and increasing lipid motion below TM [25,36,37],
but lipid unsaturation moderates this effect [25,26,29]. We
have been interested in how the presence of cholesterol in
both saturated and unsaturated membranes may change the
CPZ effect on lipid order and motion. The measurements
were carried out above and below TM of both membranes (at
37 and 15 jC in case of DMPC, and at 15 and 2 jC in case
of SOPC). Because of a possible role of the positive charge
of CPZ, we studied its interaction with the membranes at
two pH: at pH 9.0, in which CPZ is present predominantly
in neutral form and at pH 7.4 when it should be protonated
and positively charged [35].
3.2.1. Chlorpromazine in DMPC-cholesterol membranes
Fig. 2 shows the effect of 10 and 30 mol% of CPZ on
order parameter S (Fig. 2A) and Amax parameter (Fig. 2B) of
CSL incorporated into DMPC membranes containing 0 and
30 mol% of cholesterol.
DS is a difference between the S values for the mem-
branes in the presence of 10 or 30 mol% of CPZ and the
control values of S (obtained from the membranes with 0
and 30 mol% cholesterol, respectively). It is clear that for
DMPC membranes without cholesterol the ordering effect
of CPZ (above TM) becomes noticeable for the higher CPZ
concentration, and at pH 7.4. Interestingly, in the presence

Fig. 1. [DAmax] values (in gauss) of CSL, 5-SASL and 16-SASL in DMPC (black bars) and DMPG (gray bars) membranes at 15 jC, pH 9.0. DAmax was
calculated as a difference between the values of Amax in the membranes containing 10 mol% CPZ and the control values of Amax in pure membranes. The bars
indicate mean F SD of at least four independent measurements; * indicates a statistically significant difference (the Student’s t-test, p V 0.05) relative to DMPC
membrane, ** indicates a significant difference relative to CSL.
 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52 47

Fig. 2. [DS] values (A) and [DAmax] values (in gauss; B) of CSL in DMPC membranes at pH 7.4 (black bars) and pH 9.0 (gray bars). DS and DAmax are the
differences between the values of S and Amax parameters, respectively, obtained from the membranes in the presence of 10 or 30 mol% CPZ and their control
values (obtained from the membranes without cholesterol-left panel, and containing 30 mol% cholesterol-right panel of the graph). S values were obtained at
37 jC and Amax values at 15 jC. The bars indicate mean F SD, and ** indicates a statistically significant difference (the Student’s t-test, p V 0.05) relative to
10 mol% CPZ, *** indicates a significant difference relative to 0 mol% cholesterol.

of 30 mol% of cholesterol, CPZ has the opposite effect on
DMPC membrane. Instead of the slight ordering effect, a
clear increase in lipid motion has been observed at CPZ
concentration of 10 mol%. This effect was observed at both
pH with no significant difference.
Below TM, DAmax was measured as a difference between
the values of Amax for the membranes in the presence of 10
or 30 mol% of CPZ and Amax control values (obtained from
the membranes with 0 and 30 mol% cholesterol, respective-
ly). A strong increase in lipid motion was seen for pure
DMPC membranes in the presence of CPZ, at both concen-
trations and regardless of pH. In the presence of 30 mol% of
cholesterol, CPZ also increased lipid motion, but the effect
was smaller than in pure membranes.
3.2.2. Chlorpromazine in SOPC-cholesterol membranes
Similar experiments were performed on unsaturated
SOPC membranes. Fig. 3 shows the effect of 10 and 30
mol% of CPZ on order parameter S (Fig. 3A) and Amax
parameter (Fig. 3B) of CSL incorporated into SOPC mem-
branes containing 0 and 30 mol% of cholesterol. Above TM,
the ordering effect of CPZ can be seen in pure SOPC
membranes. This effect increases with CPZ concentration
and is stronger at pH 7.4 than at pH 9.0. The ordering effect
of CPZ observed in pure SOPC membrane has been
completely reduced at the presence of 30 mol% cholesterol,
but unlike in DMPC-cholesterol membrane, no fluidizing
effect has been observed instead.
Below TM, at both pH, a strong increase in lipid motion
has been seen for pure SOPC membranes in the presence of
CPZ. In the presence of 30 mol% of cholesterol, the
fluidizing effect of 10 mol% of CPZ was smaller than in
pure membranes, but still observed.
3.3. Spectrophotometric measurements of the partition of
chlorpromazine into saturated and unsaturated membranes
containing cholesterol
To better understand the cholesterol effect on CPZ
penetration into both membrane types, spectrophotometric
48 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52

Fig. 3. [DS] values (A) and [DAmax] values (in gauss; B) of CSL in SOPC membranes at pH 7.4 (black bars) and pH 9.0 (gray bars). DS and DAmax are the
differences between the values of S and Amax parameters, respectively, obtained from the membranes in the presence of 10 or 30 mol% CPZ and their control
values (obtained from the membranes without cholesterol-left panel, and containing 30 mol% cholesterol-right panel of the graph). S values were obtained at
15 jC and Amax values at 2 jC. The bars indicate mean F SD and * indicates a statistically significant difference (the Student’s t-test, p V 0.05) relative to pH
7.4, ** indicates a significant difference relative to 10 mol% CPZ, *** indicates a significant difference relative to 0 mol% cholesterol.

measurements of CPZ content in liposomes have been cholesterol affects the CPZ content in DMPC membranes
performed. The absorption spectra of 0.5 mM CPZ dis- more than in POPC. In cholesterol-containing DMPC lip-
solved in the buffer (pH 7.4) were compared with the osomes the amount of CPZ associated with the membrane is
spectra of CPZ remaining in liposome supernatants after about 30% lower than in pure DMPC liposomes, whereas in
the centrifugation at 4 jC (liposomes made of DMPC, POPC liposomes this difference is only about 15%. It is
DMPC-cholesterol, POPC, and POPC-cholesterol were
employed), and the partition of CPZ into liposomes has
Table 3
been determined. In case of DMPC, the observed partition CPZ partition into DMPC and POPC liposomes in the presence and in the
referred to the gel phase of the membrane, whereas in case absence of 5 mM cholesterol, at pH 7.4; 4 jC. Data at pH 9.0 not available
of POPC, it concerned the liquid-crystalline phase. The data due to high scattering effects of CPZ buffer solution
are presented in Table 3. It can be seen that CPZ quite well Lipid CPZ partition into liposomes (%)
incorporates into liposomes, with no significant difference DMPC 78.9 F 7.03
between both membranes. However, a direct comparison DMPC + cholesterol 53.7*F 6.91
between the gel and liquid-crystalline phases cannot be POPC 86.8 F 1.76
done, since, for experimental reasons (mentioned in Materi- POPC + cholesterol 74.0*F 0.71
als and methods), we were not able to determine the CPZ The CPZ partition into liposomes has been calculated as follows:
CPZ CPZ CPZ
100%  (A308 sup/A308 buf)*100%, where A308 sup is the absorbance
partition into the same membrane below and above TM. CPZ
value (at k = 308 nm) of CPZ in a given supernatant and A308 buf is the
Cholesterol reduces the CPZ content in the membranes of absorbance value of CPZ in the buffer. Errors have been calculated as SD,
both types, which is in agreement with the data on partition * indicates a statistically significant difference (the Student’s t-test, p V 0.05)
coefficient [12]. It is also clear from the data in Table 3 that relative to membranes without cholesterol.
 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52
possible that cholesterol in DMPC membranes makes CPZ
partition into the membrane more difficult, and in unsatu-
rated membranes, in which lipid alkyl chains are not easily
miscible with rigid cholesterol molecules [28], CPZ can be
better accommodated. Additionally, the difference in the
physical states of both membranes is important. The spec-
trophotometric measurements, however, do not allow telling
in what region of the membrane CPZ is located. We assume
that during liposome centrifugation all CPZ not associated
with the membrane remains in the buffer, and the rest can be
either inside the membrane, or at the membrane – water
interface, or even bound to the membrane surface.
4. Discussion
4.1. Comparison of the effects of chlorpromazine above and
below TM of the membranes
In the present work we have shown that the effects of
CPZ on membranes above and below the main phase
transition temperature (TM) are the opposite, namely order-
ing and disordering, respectively.
In literature, there is no consistent data on the effect of
local anesthetics (LA), tertiary amine amphipathic mole-
cules, on bilayer order at temperatures above TM. There are
some data showing that LA such as benzocaine, lidocaine
and tetracaine decrease molecular organization in EYPC
membranes [38,39]. Also, CPZ was found to increase
mobility of acyl methyl groups of phospholipids in DPPC/
PS membranes [4]. On the other hand, some reports indicate
that CPZ, cannabinol and pentobarbital increase the order of
egg PC-PA membranes [40] and tetracaine increases the
organization of micelles [41]. The fluorescence polarization
data of Hendrich et al. [42] have shown that another
phenothiazine derivative (FPhMS) makes EYPC, DMPC
and DPPC bilayers in a liquid-crystalline phase more rigid.
Also, molecular dynamics simulation data of Pasenkiewicz-
Gierula et al [43] have proved that the order of lipid chains
in POPC membranes penetrated by another LA (a carane
derivative) was higher than in the pure POPC bilayers. The
ordering effect of CPZ observed by us above TM is stronger
in SOPC membrane than in DMPC where it is nearly
negligible, especially at lower CPZ concentration. The lack
of noticeable effect of 10 mol% CPZ at 37 jC can be
explained by fast motion of lipid molecules and high fluidity
of the membrane at this temperature. It has been shown
before that lipid matrices can accommodate rather high
concentrations of LA without aggravating the disordered
motion of lipid chains. [44]. SOPC, which has much lower
TM than DMPC [31], was possible to be investigated in
liquid-crystalline phase at lower temperature (15 jC), in
which lipid chains are less mobile and ordering effects may
be better seen.
A decrease in lipid order caused by CPZ or some LA
below TM was observed by most authors [4,12,45,46].
49
We presume that the opposite effects of CPZ above and
below TM can be explained by different membrane physical
state, and, consequently, by the various location of this
molecule in membranes of different phases. There are data
suggesting that the mechanisms of LA interaction with
membranes above and below TM may be different. The
fluorescence quenching measurements of Hutterer et al. [47]
showed that the interaction of tetracaine with lipid mem-
branes above TM is mostly a partitioning process, whereas
below TM a saturable binding one. We think that this can be
valid also for CPZ. Above TM, a membrane can be pene-
trated by CPZ, whose rigid phenothiazine ring system
increases the alkyl chain packing and enhances their ex-
tended trans conformation in a manner similar to cholesterol
or polar carotenoids [33,48]. This suggestion can be sup-
ported by molecular dynamics simulation data, according to
which the alignment of POPC chain along an LA side chain
resembled that between DMPC alkyl chains and cholesterol
in DMPC-cholesterol membrane [43]. However, it has to be
pointed out that the ordering effect of CPZ is much smaller
than the one caused by cholesterol. Below TM, the mem-
brane penetration by CPZ is much more difficult [12]. CPZ
remains mostly at the membrane surface, or at the mem-
brane –water interface, where its polar groups may separate
the phospholipid headgroups and decrease the interaction
between them. This may lead to increasing mobility of alkyl
chains, which enter the gauche conformation [4]. The
comparison of CPZ effect to that of cholesterol was done
by Hendrich et al. [42] also for membranes below TM. It
seems however that the mechanisms of the action of both
compounds must be different, because cholesterol, as a
much bigger and more hydrophobic molecule, in contrast
to CPZ, intercalates into membranes also below TM.
4.2. Role of cholesterol and unsaturation
Addition of 30 mol% of cholesterol into DMPC and
SOPC membranes changes significantly the CPZ effects
both above and below TM. The most striking result is the
increase in lipid motion observed above TM in DMPC-
cholesterol membrane in the presence of 10 mol% CPZ (Fig.
2A). Similar results were obtained by Pang and Miller [40],
who showed that CPZ ordered egg-PC-PA bilayers with
cholesterol content up to 20 mol%, but disordered those
with greater than 25 mol%. This can be explained by the
well known fact that cholesterol strongly rigidifies the
DMPC membrane above TM [25,37] and changes its phase
from liquid-crystalline into less fluid, called liquid-ordered,
as illustrated by the DMPC/cholesterol phase diagram [49].
As shown above in the present work, in membranes with
ordered lipid chains CPZ may act as a fluidizer, increasing
lipid motion (Figs. 1 and 2B). In SOPC-cholesterol mem-
brane no increase in lipid motion was observed in the
presence of CPZ, but still the CPZ effect on this membrane
remained significantly different from its effect on pure
SOPC (Fig. 3A). Cholesterol seems to affect the CPZ-
50 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52
membrane interaction in both membrane types, but the
effect on saturated DMPC membrane is especially pro-
nounced. This result agrees with the known fact that
unsaturated membranes are less prone to the rigidifying
effect of cholesterol compared to saturated ones [25,26].
According to the DMPC/cholesterol phase diagram [49],
below TM cholesterol affects membrane fluidity in the
opposite way—it increases lipid motion, and the membrane
is no more in a typical gel-phase, but in a liquid-ordered
one. This may be a reason for observed diminishing of the
disordering effect of CPZ in both saturated and unsaturated
membranes containing cholesterol (Figs. 2B and 3B). The
lower disordering effect of CPZ in the presence of choles-
terol may also be caused by lower CPZ concentration in a
cholesterol-containing membrane. Luxnat and Galla [12]
showed that cholesterol reduced CPZ partition coefficient
into DMPC membrane quite significantly. Also, our spec-
trophotometric measurements of CPZ partition into lip-
osomes demonstrated a lower CPZ content in cholesterol-
containing membranes compared to the pure ones (Table 3).
However, the disordering effect observed in DMPC-choles-
terol membrane above TM does not allow believing that CPZ
is not present in the membrane, at least at the region close to
the polar head groups. It seems more probable that for
observed effects and differences between both cholesterol
containing membranes their different phases are responsible.
4.3. Role of pH
Chlorpromazine is a basic molecule with pK = 8.6 (in
water) [7] and 8.0 (in the presence of zwitterionic phospho-
lipid bilayers) [35]. Therefore, at pH 9.0 it is predominantly
unprotonated and neutral, and in pH 7.4-protonated with a
positive charge. Our results obtained for these two pH
values show that different forms of CPZ do not influence
the membranes equally.
Above TM, the ordering effect of CPZ on pure SOPC
membrane is stronger at pH 7.4 than at pH 9.0 (Fig. 3A). We
suppose that for DMPC membrane the dependence should
be similar, but because the ordering effect of CPZ in this
membrane is in general very small and calculated errors
comparably big, therefore we cannot draw a solid conclu-
sion about the pH dependent CPZ effect on DMPC. Any-
how, bearing in mind the limitations of the ESR method at
this point, we suggest two possible explanations for ob-
served differences. The most obvious explanation may be
that the amount of membrane associated CPZ is pH depen-
dent. Unfortunately, our spectrophotometric measurements
of CPZ partition into membranes did not give an answer at
this point, because high scattering effects at pH 9.0 made the
absorption measurements not reliable. Since CPZ is known
to affect the membrane phase transition (causing a slight
shift of TM and decrease in cooperative unit) dependently on
its concentration [50], it may be helpful to use DSC as a
highly sensitive method to determine the amount of mem-
brane associated CPZ. However, although a stronger CPZ
effect at pH 7.4 than at pH 9.0 could suggest more
membrane associated CPZ at pH 7.4, we cannot draw the
same conclusion from the data obtained below TM, where
we do not see differences in CPZ effects at both pH.
Another possible interpretation is that, due to bearing a
positive charge at pH 7.4, a CPZ molecule cannot easily
penetrate the membrane, and remains mainly at the interface
or at the bilayer surface. The charged amine group of CPZ
should be located around the phosphate atom of PC and the
most perturbing phenothiazine ring system may go deeper
into the interface region. Such a localization of CPZ affects
mostly this region of the membrane where the CSL free
radical moiety is placed. On the contrary, at pH 9.0 CPZ can
easier penetrate the membrane because of losing its positive
charge. Therefore, it can be expected that in this case CPZ is
more evenly distributed within the membrane and affects
other regions as well. Also, the molecular dynamics simu-
lation data confirm that carane derivative distribution in the
membrane depends on its charge. A neutral form of LA was
shown to cross the interfacial region and insert some frag-
ments into the bilayer core, whereas the protonated LA
molecules did not cross the membrane – water interface [43].
This interpretation is additionally supported by the data
obtained below TM, where pH does not seem to play a role
in CPZ effect on both membranes (Figs. 2B and 3B). As
mentioned previously, in the gel-phase membrane CPZ
remains mainly at the surface or at the membrane – water
interface, and cannot penetrate the membrane neither in
neutral nor in a positively charged form.
5. Conclusions
In conclusion, it should be emphasized that CPZ affects
lipid order and motion in all investigated membranes and its
effects depend first of all on the physical state of the
membranes. In saturated membranes, the slight ordering
effect observed above TM becomes a fluidizing one below
TM or in the presence of cholesterol. Cholesterol also
reduces the CPZ partition into such membranes. In unsatu-
rated membranes, cholesterol influences the CPZ content in
the membrane and its effects to less extent. The pH, which
affects the CPZ charge, is also important, but only for pure
membranes in the liquid-crystalline phase.
For further photochemical study of CPZ as a sensitizer,
liposomes made of unsaturated lipids and containing cho-
lesterol are needed to be employed. However, cholesterol as
a molecule altering the state of membranes and the effect of
CPZ so strongly, does not seem to be a good probe for such
tests as iodometric assay and a HPLC-EC(Hg), where the
low temperature of measurements is required. At low tem-
perature, saturated membranes are usually in the gel phase,
whereas unsaturated in the liquid-crystalline one. It has to be
pointed out that not only the localization and, consequently,
the effectiveness of CPZ can be different in both membrane
types, but also the addition of cholesterol may influence the
 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52
CPZ amount and effects in both membranes in a different
way. Therefore, for photochemical studies, we recommend
using lower amounts of cholesterol than 30 mol%. It has to
be underlined that comparison of the results from different
model membranes requires a very careful analysis of exper-
imental conditions, such as temperature, pH and membrane
composition, especially cholesterol content.
Acknowledgements
We thank Prof. Tadeusz Sarna for providing us with
SOPC and DMPG Prof. Gerard Beijersbergen van Hene-
gouwen for providing us with CPZ, and Dr. Tomasz Róg for
fruitful discussion.
References
[1] G. Housley, G.V. Born, D.M. Conroy, D.M.J. Berlin, A.D. Smith,
Influence of dietary lipids on the effect of chlorpromazine on mem-
brane properties of rabbit red cells, Proc. R. Soc. Lond., B Biol. Sci.
227 (1986) 43 – 51.
[2] D.N. Brindley, M. Bowley, Drugs affecting the synthesis of glyc-
erides and phospholipids in rat liver: the effects of clofibrate,
halofenate, fenfluramine, amphetamine, cinchocaine, chlorproma-
zine, demethylimipramine, mepyramine and some of their deriva-
tives, J. Biochem. 148 (1975) 461 – 469.
[3] F.P. Bymaster, L.D. Nelson, N.W. Delapp, J.F. Falcone, K. Eckols,
L.L. Truex, M.M. Foremann, V.L. Lucaites, D.O. Calligaro, Antago-
nism by olanzapine of dopamine D1, serotonin2, muscarinic, hista-
mine H1 and a1-adrenergic receptors in vitro, Schizophr. Res. 37
(1999) 107 – 122.
[4] W. Nerdal, S.A. Gundersen, V. Thorsen, H. Hoiland, H. Holmsen,
Chlorpromazine interaction with glycerophospholipid liposomes stud-
ies by magic angle spinning solid state 13 C-NMR and differential
scanning calorimetry, Biochim. Biophys. Acta 1464 (2000) 165 – 175.
[5] A.V. Agasøsler, H. Holmsen, Chlorpromazine associates with phos-
phatidylserines to cause an increase in the lipid’s own interfacial
molecular area-role of fatty acyl composition, Biophys. Chem. 91
(2001) 37 – 47.
[6] A.V. Agasøsler, L.M. Tungodden, D. Cejka, E. Bakstad, L.K. Sydnes,
H. Holmsen, Chlorpromazine-induced increase in dipalmitoylphos-
phatidylserine surface area in monolayers at room temperature, Bio-
chem. Pharmacol. 61 (2001) 817 – 825.
[7] H. Ahyayauch, F.M. Goni, M. Bennouna, pH-dependent effects of
chlorpromazine on liposomes and erythrocyte membranes, J. Lipo-
some Res. 13 (2003) 147 – 155.
[8] J. Frenzel, K. Arnold, P. Nuhn, Calorimetric, 13C NMR, and 31P NMR
studies on the interaction of some phenothiazine derivatives with
dipalmitoyl phosphatidylcholine model membranes, Biochim. Bio-
phys. Acta 507 (1978) 185 – 197.
[9] S.R.W. Louro, C. Anteneodo, E. Wajnberg, Carboxyl groups at the
membrane interface as molecular targets for local anesthetics, Bio-
phys. Chem. 74 (1998) 35 – 43.
[10] J. Römer, M.H. Bickel, Interactions of chlorpromazine and imipra-
mine with artificial membranes investigated by equilibrium dialysis,
dual-wavelength photometry, and fluorimetry, Biochem. Pharmacol.
28 (1979) 799 – 805.
[11] A. Jutila, T. Söderlund, A.L. Pakkanen, M. Huttunen, P.K.J. Kinnu-
nen, Comparison of the effects of clozapine, chlorpromazine, and
haloperidol on membrane lateral heterogeneity, Chem. Phys. Lipids
112 (2001) 151 – 163.
51
[12] M. Luxnat, H.-J. Galla, Partition of chlorpromazine into lipid bilayer
membranes: the effect of membrane structure and composition, Bio-
chim. Biophys. Acta 856 (1986) 274 – 283.
[13] S. Takegami, K. Kitamura, T. Kitade, K. Hasegawa, A. Nishihira,
Effects of particle size and cholesterol content on the partition coef-
ficients of chlorpromazine and triflupromarizne between phosphati-
dylcholine-cholesterol bilayers of unilammellar vesicles and water
studied by second-derivative spectrophotometry, J. Colloid Interface
Sci. 220 (1999) 81 – 87.
[14] T.B. Fritzpatrick, M.A. Pathak, I.A. Magnus, W.L. Curwen, Abnor-
mal reactions of man to light, Annu. Rev. Med. 14 (1963) 195 – 214.
[15] A.S. Zelickson, H.C. Zeller, New and unusual reaction to chlorprom-
azine, J. Am. Med. Assoc. 188 (1964) 394 – 396.
[16] B. Ljunggren, G. Bojs, A case of photosensitivity and contact allergy
to systemic tricyclic drugs, with unsual features, Contact Dermatitis
24 (1991) 259 – 265.
[17] A.G. Motten, G.R. Buettner, C.F. Chignell, Spectroscopic studies of
cutaneus photosensitizing agents-VIII. A spin trapping study of light
induced free radicals from chlorpromazine and promazine, Photo-
chem. Photobiol. 42 (1985) 9 – 15.
[18] E. Gocke, Review of the genotoxic properties of chlorpromazine and
related phenothiazines, Mutat. Res. 366 (1996) 9 – 21.
[19] B. Ljunggren, H. Moller, Phenothiazine phototoxicity: an experimen-
tal study on chlorpromazine and its metabolites, J. Invest. Dermatol.
68 (1977) 313 – 317.
[20] S.A. Schoonderwoerd, G.J.M. Beijersbergen van Henegouwen, S.
Van Belkum, In vivo photodegradation of chlorpromazine, Photo-
chem. Photobiol. 5 (1989) 659 – 664.
[21] R.D. Hall, G.R. Buettner, A.G. Motten, C.F. Chignell, Near-infrared
detection of singlet molecular oxygen produced by photosensitization
with promazine and chlorpromazine, Photochem. Photobiol. 46
(1987) 295 – 300.
[22] W. Korytowski, G.J. Bachowski, A.W. Girotti, Chromatographic sep-
aration and electrochemical determination of cholesterol hydroper-
oxides generated by photodynamic action, Methods Enzymol. 197
(1991) 149 – 156.
[23] W. Korytowski, A.W. Girotti, Singlet oxygen adducts of cholesterol:
photogeneration and reductive turnover in membrane systems, Photo-
chem. Photobiol. 70 (1999) 484 – 489.
[24] A.W. Girotti, W. Korytowski, Cholesterol as a singlet oxygen detector
in biological systems, Methods Enzymol. 319 (2000) 85 – 100.
[25] A. Kusumi, W.K. Subczynski, M. Pasenkiewicz-Gierula, J.S. Hyde,
H. Merkle, Spin-label study on phosphatidylcholine-cholesterol mem-
branes: effect on alkyl chain length and unsaturation in the fluid
phase, Biochim. Biophys. Acta 854 (1986) 307 – 317.
[26] M. Pasenkiewicz-Gierula, W.K. Subczynski, A. Kusumi, Rotational
diffusion of a steroid molecule in phosphatidylcholine-cholesterol
membranes: fluid-phase micro-immiscibility in unsaturated phos-
phatidylcholine-cholesterol membranes, Biochemistry 29 (1990)
4059 – 4069.
[27] W.K. Subczynski, A. Wisniewska, J.-J. Yin, J.S. Hyde, A. Kusumi,
Hydrophobic barriers of lipid bilayer membranes formed by reduction
of water penetration by alkyl chain unsaturation and cholesterol, Bio-
chemistry 33 (1994) 7670 – 7681.
[28] A. Wisniewska, J. Draus, W.K. Subczynski, Is a fluid-mosaic model
of biological membranes fully relevant? Studies on lipid organization
in model and biological membranes, Cell. Mol. Biol. Lett. 8 (2003)
147 – 159.
[29] M. Pasenkiewicz-Gierula, W.K. Subczynski, A. Kusumi, Influence of
phospholipid unsaturation on the cholesterol distribution in mem-
branes, Biochimie 73 (1991) 1311 – 1316.
[30] A. Pezeshk, V. Pezeshk, A. Firlej, J. Wojas, W.K. Subczynski, Trans-
port of spin-labeled tetracycline across model and biological mem-
branes, Life Sci. 52 (1993) 1071 – 1078.
[31] M.W. Hawco, P.J. Davis, K.M. Keough, Lipid fluidity in lung sur-
factant: monolayers of saturated and unsaturated lecithins, J. Appl.
Physiol. 51 (1981) 509 – 515.
52 A. Wisniewska, A. Wolnicka-Glubisz / Biophysical Chemistry 111 (2004) 43–52
[32] D. Marsh, Electron spin resonance: spin labels, in: E. Grell (Ed.), Mem-
brane Spectroscopy, Springer-Verlag, New York, 1981, pp. 51 – 142.
[33] W.K. Subczynski, E. Markowska, W.I. Gruszecki, J. Sielewiesiuk,
Effect of polar carotenoids on dimyristoylphosphatidylcholine mem-
branes: a spin-label study, Biochim. Biophys. Acta 1105 (1992)
97 – 108.
[34] L.J. Berliner, Spin labeling in enzymology: spin-labeled enzymes and
proteins, Methods Enzymol. 49 (1978) 418 – 480.
[35] J.Y. Chen, L.S. Brunauer, F.C. Chu, C.M. Helsel, M.M. Gedde, W.H.
Huestis, Selective amphipathic nature of chlorpromazine binding to
plasma membrane bilayers, Biochim. Biophys. Acta 1616 (2003)
95 – 105.
[36] W.K. Subczynski, A. Kusumi, Effects of very small amounts of cho-
lesterol on gel-phase phosphatidylcholine membranes, Biochim. Bio-
phys. Acta 854 (1986) 318 – 320.
[37] T.P. McMullen, R.N. Lewis, R.N. McElhaney, Differential scanning
calorimetric study of the effect of cholesterol on the thermotropic
phase behavior of a homologous series of linear saturated phosphati-
dylcholines, Biochemistry 32 (1993) 516 – 522.
[38] L.M. Pinto, D.K. Yokaichiya, L.F. Fraceto, E. de Paula, Interaction
of benzocaine with model membranes, Biophys. Chem. 87 (2000)
213 – 223.
[39] L.F. Fraceto, L.M.A. Pinto, L. Franzoni, A.A.C. Braga, A. Spisni, S.
Schreier, E. de Paula, Spectroscopic evidence for a preferential loca-
tion of lidocaine inside phospholipid bilayers, Biophys. Chem. 99
(2002) 229 – 243.
[40] K.Y. Pang, K.W. Miller, Cholesterol modulates the effects of mem-
brane perturbers in phospholipid vesicles and biomembranes, Bio-
chim. Biophys. Acta 511 (1978) 1 – 9.
[41] C.V. Teixeira, R. Itri, F. Casallanovo, S. Schreier, Local anesthetic-
induced microscopic and mesoscopic effects in micelles. A fluores-
cence, spin label and SASXS study: single angle X-ray scattering,
Biochim. Biophys. Acta 1510 (2001) 93 – 105.
[42] A.B. Hendrich, O. Wesolowska, M. Komorowska, N. Motohashi, K.
Michalak, The alterations of lipid bilayer fluidity induced by newly
synthesized phenothiazine derivative, Biophys. Chem. 98 (2002)
275 – 285.
[43] M. Pasenkiewicz-Gierula, T. Róg, J. Grochowski, P. Serda, R. Czar-
necki, T. Librowski, S. Lochyński, Effects of a carane derivative local
anesthetic on a phospholipid bilayer studied by molecular dynamics
simulation, Biophys. J. 85 (2003) 1 – 11.
[44] F. Momo, S. Fabris, A. Bindoli, G. Scutari, R. Stevanato, Different
effects of propofol and nitroso propofol on DMPC multilamellar lip-
osomes, Biophys. Chem. 95 (2002) 145 – 155.
[45] A.B. Hendrich, O. Wesolowska, K. Michalak, Trifluoperazine indu-
ces domain formation in zwitterionic phosphatidylcholine but not in
charged phosphatidylglycerol bilayers, Biochim. Biophys. Acta 1510
(2001) 414 – 425.
[46] F. Momo, S. Fabris, A. Wisniewska, C. Fiore, A. Bindoli, G. Scutari,
R. Stevanato, Correlation between fluidising effects on phospholipid
membranes and mitochondrial respiration of propofol and p-nitroso-
phenol homologues, Biophys. Chem. 103 (2003) 213 – 222.
[47] R. Hutterer, K. Krämer, F.W. Schneider, M. Hof, The localization of
the local anesthetic tetracaine in phospholipid vesicles: a fluorescence
quenching and resonance energy transfer study, Chem. Phys. Lipids
90 (1997) 11 – 23.
[48] W.K. Subczynski, A. Wisniewska, Effect of polar carotenoids on the
shape of the hydrophobic barrier of phospholipid bilayers, Biochim.
Biophys. Acta 1368 (1998) 235 – 246.
[49] M.B. Sankaram, T.E. Thompson, Cholesterol-induced fluid-phase im-
miscibility in membranes, Proc. Natl. Acad. Sci. U. S. A. 88 (1991)
8686 – 8690.
[50] P.P. Constantinides, N. Inouchi, T.R. Tritton, A.C. Sartorelli, J.M.
Sturtevant, A scanning calorimetric study of the interaction of anthra-
cyclines with neutral and acidic phospholipids alone and in binary
mixtures, J. Biol. Chem. 261 (1986) 10196 – 10203.